Why LINCS?

The core of the Linked Infrastructure for Networked Cultural Scholarship (LINCS) project is the simple proposition that making Linked Open Data (LOD) out of the stuff scholars use to understand and analyze culture will make a difference. LINCS hopes to make a difference to how we can make sense of the human past and present. It aims to enable such work within and beyond academic contexts, and in so doing to improve how things cultural can be presented and circulated on the World Wide Web.

I say stuff and things to demystify the notion of data. Those who make culture seldom think of themselves as creating data, and neither do those who collect, celebrate, curate, and analyze cultural objects, processes, and events. But everything we make and do can be represented as data and circulated on the Web, and a large part of contemporary culture is Web culture...

A sense of opposition between culture and technology has a long history (Keep, 2003), and there are good reasons to challenge simplistic notions of data with the idea of capta, which recognizes that all data is mediated and situated (Drucker, 2011). Numerous studies demonstrate how technology can perpetuate past and present wrongs and underscore the perils of ignoring the human and cultural implications of technology (see Chun, 2016; Nakamura, 2002; Noble, 2018). The COVID-19 pandemic and the #BlackLivesMatter movement have made it clear that participatory democracy requires good data. Because the production and impact of data are massively affected by cultural contexts and histories (Wernimont, 2019), it is all the more crucial to aggregate online information in ways that are transparent, responsible, and trustworthy.

Linked Open Data uses web technologies and standards to weave meaning into the fabric of the Web itself. It connects related stuff from different websites so that humans and machines can talk about the same things in the same terms, and can advance knowledge and debate about those things collaboratively.

Later posts, workshops, and events will explore Linked Open Data and how it works more fully. For now, I will just note that LINCS is motivated by the promise that Linked Data (LD) can help to:

• break down silos separating material on the same topic from different websites
• enrich and contextualize web content from sources across the Web
• make it easier to find content and zero in on what is most relevant
• analyze content from different sources together
• promote sharing and reuse, and enable collective knowledge production
• reveal connections between things through visualization
• support better queries to answer more complex questions of Web content
• enable inferences and so add up to more than the sum of its parts

tip

New to Linked Data?

This 4-minute video from Europeana provides a great overview. For a researcher perspective, check out this Programming Historian introduction by Jonathan Blaney. If you’re after quick definitions for terms such as ontology, check out the LINCS Glossary in progress.

To see Linked Open Data in action, check out interviews with legendary musicians on Linked Jazz, profiles of Virginia Woolf or Douglas Adams pulled from multiple sources, a gathering of works of art, or explore Wikidata using the Metaphactory demo.

Linked Open Data done right helps build the Semantic Web (see Shadbolt, Hall, & Berners-Lee, 2006). But although based on open standards and intended for the public good, Linked Data> is increasingly driven by commercial interests. Uber, Facebook, and the for–profit publisher Elsevier, for instance, have huge Linked Data sets, or what are now often called knowledge graphs. Results generated by Google’s proprietary knowledge graph, drawing in much open Linked Data, appear in the boxes on the right of many search result screens.

LINCS responds to the need to create linked web content that serves academic and public rather than corporate interests, and it recognizes that the complexity of the technologies involved means that this cannot happen without a shared, open infrastructure. LINCS is a Cyberinfrastructure project funded by the Canada Foundation for Innovation, provincial governments, universities, and other partners to convert existing cultural and scholarly materials from across many disciplines in the humanities into Linked Data, and to provide documentation and training so that researchers can use, enhance, and create more such data to support future work.

LINCS began officially in April of 2020. Its phased approach starts with datasets that are closer in structure to Linked Data, such as spreadsheets and databases, and works its way towards more complex ones such as those involving natural language. Decisions about how to build the infrastructure are proceeding in dialogue with our community, thanks to an active Research Board. We are currently interviewing scholars and forming interest groups, and our online presentations and workshops on everything from basic introductions to technological choices, tools, and methods will be open to all.

LINCS is an open project. If you’re a researcher who would like to learn more about Linked Data, mobilize your own dataset, or test things; a designer or developer interested in joining designfests or hackfests; or you represent a project or institution that wants to connect with LINCS as an organization, please be in touch. Sign up for our newsletter, request an invitation to our Slack announcements channel, or follow our GitLab repository to keep on top of new developments.

---

“Why LINCS?” can be answered quite differently, however, by invoking a colleague rather than a concept.

Stéfan Sinclair, the LINCS site lead at McGill.

As many in the DH community worldwide mourn our loss, I want to outline how deeply this project is and will always be indebted to him.

The germ in many ways goes back to a course in Digital Tools for Literary History that Stan Ruecker and I taught in 2009 at DHSI. Participants explored and discussed a number of approaches and tools, including an accessible new browser–based text analysis tool called “Voyeur.” It was super easy to get going with and gave immediate results. To say that it wowed the group is an understatement: it was the star of the course. I had by then sampled the heady brew of iterative design and development through collaboration with Stéfan, Stan, and Milena Radzikowska on tools such as the Dynamic Table of Contexts and the Mandala browser. But that course brought home to me the power and potential of accessible, web–based DH tools.

That platform became a model of what the trifecta of creation, collaboration, and community can achieve. It was grounded in the enduring intellectual partnership between Stéfan and Geoffrey Rockwell, the joy of which I witnessed on countless occasions, most notably perhaps in the dialogue they performed at DH2007 at UIUC (see Rockwell & Sinclair, 2007). And that collaborative dialogue extended outwards to all who took an interest in it. As we began exploring how we might build a bridge between their tool and the newly funded Canadian Writing Research Collaboratory (CWRC), I recall sitting in Edmonton’s LEVA Café and raising the concern that if they wanted women to use the tool, there were problems with the connotations of its name as well as the results of Googling “Voyeur.” Despite their fondness for a name that was already something of a brand, it was in short order changed to “Voyant.” That kind of listening is rare.

It has been a privilege to watch Voyant become the go–to tool for DH teaching and text exploration, the most–used tool on Compute Canada’s Cloud. Voyant is a “labour of love” between two collaborators—an evolving means of thinking through text technologies for its creators that enabled others to do the same. The progeny of Stéfan’s fascination with the OuLiPo movement, Voyant boasts tools that include Bubbles’ ebullient twist on datafication and Bubblelines’ strange beauty. After all, he said, “tools are playgrounds that structure scenes for interpretation” (see Rockwell, 2007).

When the Cyberinfrastructure program was announced and the Canadian DH community began to consult with each other, Stéfan and I were both thinking of applying. With typical generosity, he suggested that we combine on a single application from McGill, and then various factors led to applying from Guelph instead. The focus shifted as we developed a first and then a second application, but Stéfan has always been at the heart of this project’s vision of how digital infrastructure must be grounded in intellectual engagement and community, and tools fully integrated into “the research cycle” (Rockwell & Sinclair, 2022).

As Voyant’s lead developer, Stéfan was infrastructure personified: understated, omnipresent, essential. Engaged in service and support for research in ways that often go under–recognized, he infused Voyant’s code with his astute grasp of the dialectical relationship between how we shape tools and how tools shape us. The tributes to Stéfan that are now pouring in make abundantly clear his brilliant, multifaceted, and diverse contributions to the Digital Humanities (DH). He demonstrated a commitment to community at every level, from warmly drawing conference newcomers into the fold and mentoring students near and far, through both leading prominently and shepherding from behind, to the innovation, care, and repair of many infrastructures including but extending well beyond Voyant.

LINCS will be the poorer for the loss of Stéfan, but what he did and who he was will always be at this project’s heart and its continuing inspiration.

note

Special thanks to Geoffrey for posting the dialogue “Reading Tools, or Text Analysis Tools as Objects of Interpretation”, presented with Stéfan at DH2007.

---

Works Cited

Chun, Wendy Hui Kyong. Updating to Remain the Same: Habitual New Media. MIT Press, 2016.

Drucker, Johanna. “Humanities Approaches to Graphical Display.” Digital Humanities Quarterly 5, no. 1 (2011): 1–22. https://dhq-static.digitalhumanities.org/pdf/000091.pdf.

Keep, Christopher. “Of Writing Machines and Scholar–Gipsies.” ESC 29, no. 1–2 (2003): 55–66. https://ojs.lib.uwo.ca/index.php/esc/article/view/10181/8278.

Nakamura, Lisa. Cybertypes: Race, Ethnicity, and Identity on the Internet. Routledge, 2002.

Noble, Safiya Umoja. Algorithms of Oppression: How Search Engines Reinforce Racism. New York Univeristy Press, 2018.

Rockwell, Geoffrey. “Celebrating Stéfan Sinclair: A Dialogue from 2007.” Theoreti.ca. August 11, 2020. https://theoreti.ca/?p=7599

Rockwell, Geoffrey & Stéfan Sinclair. Hermeneutica: Computer–Assisted Interpretation in the Humanities. MIT Press, 2022. https://mitpress.mit.edu/9780262545891/

Rockwell, Geoffrey & Stéfan Sinclair. “Reading Tools, or Text Analysis Tools as Objects of Interpretation - Dialogue with Stéfan Sinclair and Geoffrey Rockwell.” Script presented at Digital Humanities. University of Illinois, Urbana–Champaign. June 2007. https://doi.org/10.7939/r3-9vkj-k220.

Shadbolt, Nigel, Wendy Hall, & Tim Berners–Lee. “The Semantic Web Revisited.” IEEE Computer Society (2006): 96–101. https://eprints.soton.ac.uk/262614/1/Semantic_Web_Revisted.pdf.

Wernimont, Jacqueline. Numbered Lives: Life and Death in Quantum Media. MIT Press, 2019. https://mitpress.mit.edu/books/numbered-lives.